Yay!!! Google is finally in the arena.
They just announced Gemini, and it has some impressive benchmark scores ππ
Gemini beats GPT4 in several benchmarks except HellaSwag. Interestingly, top open-source models score around 87-88 on HellaSwag. The 4-pt MMLU beat isβ¦
@alignment_lab
It will be interesting to see if synthetic data from advanced models like Gemini makes its way into open source models.
More interesting also to see if the time from model release to open source "clone" gets reduced over time.